Mortality from lung cancer could be reduced substantially through early detection and the implementation of targeted approaches for chemoprevention and treatment of early stage cancers. We conducted the first study in collaboration with the Colorado Lung SPORE to prospectively evaluate methylation in sputum of a large panel of genes for their ability to predict lung cancer. This nested, case-control study of persons from the Colorado Cohort revealed that a panel of genes could predict incident lung cancer between 3 and 18 months prior to clinical diagnosis with both a sensitivity and specificity of 64%. This same marker panel is now being used to evaluate methylation in sputum from prevalent stage I lung cancer cases compared to a second cohort of high-risk smokers. Methylation of three or more genes of a seven-gene panel revealed a sensitivity of 75% and a specificity of 81%. Evaluation of additional candidate biomarkers has identified other promising genes for inclusion in the ultimate panel for early detection of lung cancer. These studies support advancing a methylation gene panel to full clinical validation as a tool for early detection. This will be accomplished through collaboration with the Colorado SPORE by first determining the optimal gene panel for distinguishing newly diagnosed stage I lung cancer from cancer-free smokers. We will then validate the performance of the ultimate gene panel for early detection of lung cancer using ACRIN, a prospective cohort of people at high risk for lung cancer within the National Lung Screening Trial who are randomized to lung cancer screening modalities. Through a nested, case-control design, the sensitivity, specificity, positive, and negative predictive values of this gene panel will be determined. A second major goal for this project is to collaborate closely with Project 1 in which the reversal of abnormal gene silencing is being tested as a therapeutic target for lung cancer. The strategy of using the demethylating agent 5-azacytidine and the histone deacetylase inhibitors, sodium phenylbutyrate or MS275 has shown promising responses in the treatment of myeloid malignancies. This therapy may also be effective in an adjuvant setting to prevent recurrence of cancer in resected stage I lung cancer patients. Through collaboration with Project 1, we will use gene methylation in sputum as a biomarker to predict therapeutic response and recurrence of lung cancer in resected stage I lung cancer patients receiving adjuvant treatment with demethylating agents. Relevance to Public Health: These studies should ultimately lead to the development of an approved diagnostic test for early detection of lung cancer. In addition, these studies should clearly impact how early stage lung cancer is managed by creating a molecular-based test to guide treatment decisions and provide biomarkers for monitoring the efficacy of demethylation therapy.